We have found in the fear potentiated startle test that bilateral local infusion of morphine into the medial nucleus of the amygdala (Me) completely blocks the expression of conditioned fear whereas infusion of comparable doses into the central (Ce), basolateral (Bla) or caudate-putamen had no effect. The effects of morphine infused into the Me were completely reversed by co-infusion of naloxone into the Me. Moreover, local infusion of naloxone into the Me completely prevented morphine, given systemically, from blocking fear potentiated startle. Local infusion of mu or delta selective opiate agonists had similar effects to morphine. At the present time it is not clear why local infusion of morphine into the Me so effectively blocks fear potentiated startle because we have not found that lesions of the Me block fear potentiated startle. Although the Me has been reported to project to the Ce we believe this apparent projection probably resulted from spread of the tracer into the Ce, because we have not found projections from the Me to the Ce using discrete infusion of the sensitive anterograde tracer biotinylated dextran into the Me. Instead, the Me projects heavily to anteriomedial bed nucleus of the stria terminalis (BNSTam) the posterior intermediate BNSTpi), and the ventromedial hypothalamus (VMH). Because the BNST projects heavily to the Ce, in fact most heavily to the output division of the Ce projections from the Me to the BNST and then to the Ce could well be involved in an inhibitory circuit activated by morphine binding to receptors in the Me. Alternatively, other brain areas might also participate in this putative fear- inhibition circuit. To evaluate this, we infused morphine into the Me and then analyzed the brain using immunocytochemistry for Fos protein as a marker of neuronal activation. Local infusion of morphine into the ME led to increases in Fos in the Me, ventral lateral septal nucleus and adjacent areas of the BNST, the intercalated cell mass on the border between the Ce and Bla, and the ventral subiculum. Because some of these areas have been implicated in the inhibition of fear we believe we may have tapped into a neuronal network involved in the inhibition of fear. Hence, we will determine the role of the BNST, ventral subiculum, lateral septum, ventromedial hypothalamus and intercalated cells in the anxiolytic effects of morphine given locally into the Me, as well as the anxiolytic effects of morphine, buspirone and diazepam given systemically. We wiU also evaluate the role of opiate receptors in the Me, neuropeptide Y (NPY) cells in the Bla and GABA in the Bla, as well as various brain areas identified above, in extinction and conditioned inhibition of fear potentiated startle. Finally, we will carry out brain imaging of odor-mediated fear retention, extinction, and conditioned inhibition in anesthetized rats usingfMRI in an 11 T magnet measured with blood pressure in the magnet and fear potentiated startle in later tests in unanesthetized rats.